ABSTRACT There is currently no recognized way of accurately predicting who will recover from post-traumatic headache (PTH) during the acute phase following concussion and who will go on to develop persistent post-traumatic headache (PPTH), a condition that is difficult to treat effectively. Clinical experience with treating secondary headaches suggests that treatment is most effective when initiated early after headache onset, before headache patterns become persistent. So, why don't clinicians treat all patients with PTH acutely following concussion? Although it is likely to be beneficial to treat those patients who are at high risk for PPTH as early as possible, medication side effects and toxicities make it potentially harmful to administer unnecessary medication to those patients who would recover on their own without treatment. The inability to identify which concussed patients with acute PTH are at high risk for PTH persistence leaves clinicians without the knowledge needed to make informed decisions regarding how aggressive to manage patients with acute PTH. Study Goal: to develop a prognostic biomarker signature for PPTH using clinical data as well as structural and functional brain neuroimaging and to assess the predictive accuracy of an ensemble biomarker signature for the early identification of patients at high risk for PTH persistence. The accurate prediction of individuals who are at high risk for PTH persistence would allow clinicians to recommend early and more aggressive management with the intention of preventing PTH persistence. A predictive model for PPTH could also guide clinical trials that will test the utility of innovative therapies to prevent PPTH by allowing for enrichment of the subject cohort with patients at high risk of PPTH. R61 Phase: During the R61 phase this study will develop brain imaging and clinical feature prognostic biomarker signatures for PPTH using machine-learning algorithms. The imaging and clinical biomarker signatures will be developed by laboratories that will first work in-parallel, and then the laboratories will combine the neuroimaging and clinical biomarkers using an ensemble approach. Results of this study will determine important clinical factors (e.g. demographics, medical history, brain injury characteristics, headache characteristics, speech patterns) and brain imaging markers (e.g. regional volumes, cortical thickness, white matter tract integrity, perfusion, functional connectivity) for predicting PTH persistence Additionally, the predictive weight of specific clinical factors and neuroimaging features for characterizing patients who are at higher risk for developing PPTH will be determined. R33 Phase: Once the predictive weights of clinical factors and neuroimaging features are determined, the clinical testing battery and neuroimaging sequences can be pruned down and optimized to include only those that have high predictive power for PPTH. The scanning sequences used in this study are commonly available MRI sequences, making their use feasible across healthcare centers. This optimized and shortened testing sequence can be translated into clinical practice and integrated into PTH clinical trials for early identification of those individuals who are at high risk for PTH persistence.